def test_html_soup(self):
element_walker = ElementWalker(css_scraper=CSSScraper(),
javascript_scraper=JavaScriptScraper())
scraper = HTMLScraper(HTMLParser(), element_walker)
request = Request('http://example.com/')
response = Response(200, '')
response.body = Body()
response.fields['Refresh'] = 'yes'
with wpull.util.reset_file_offset(response.body):
html_file_path = os.path.join(ROOT_PATH, 'testing', 'samples',
'soup.html')
with open(html_file_path, 'rb') as in_file:
shutil.copyfileobj(in_file, response.body)
scrape_result = scraper.scrape(request, response)
inline_urls = scrape_result.inline_links
linked_urls = scrape_result.linked_links
self.assertEqual({'http://example.com/ABOUTM~1.JPG'}, inline_urls)
self.assertEqual(
{
'http://example.com/BLOG',
'http://example.com/web ring/Join.htm',
}, linked_urls)
def test_rss_as_html(self):
element_walker = ElementWalker(css_scraper=CSSScraper(),
javascript_scraper=JavaScriptScraper())
scraper = HTMLScraper(HTMLParser(), element_walker)
request = Request('http://example.com/')
response = Response(200, '')
response.body = Body()
response.fields['content-type'] = 'application/rss+xml'
with wpull.util.reset_file_offset(response.body):
html_file_path = os.path.join(ROOT_PATH, 'testing', 'samples',
'rss.xml')
with open(html_file_path, 'rb') as in_file:
shutil.copyfileobj(in_file, response.body)
scrape_result = scraper.scrape(request, response)
self.assertTrue(scrape_result)
inline_urls = scrape_result.inline_links
linked_urls = scrape_result.linked_links
self.assertFalse(inline_urls)
self.assertEqual(
{
'http://www.someexamplerssdomain.com/main.html',
'http://www.wikipedia.org/'
}, linked_urls)
def test_sitemap_encoding(self):
parser = HTMLParser()
reader = SitemapReader(parser)
bom_map = {
'utf_16_le': codecs.BOM_UTF16_LE,
'utf_16_be': codecs.BOM_UTF16_BE,
'utf_32_le': codecs.BOM_UTF32_LE,
'utf_32_be': codecs.BOM_UTF32_BE,
}
for name in CODEC_NAMES:
if name in EBCDIC or name == 'utf_8_sig':
# XXX: we're assuming that all codecs are ASCII backward
# compatable
continue
if name.startswith('utf_16') or name.startswith('utf_32'):
# FIXME: libxml/lxml doesn't like it when we pass in a codec
# name but don't specify the endian but BOM is included
continue
data = io.BytesIO(
bom_map.get(name, b'') +
'<?xml version="1.0" encoding="UTF-8"?>'
'<urlset><url><loc>blah</loc></url></urlset>'.encode(name)
)
print('->', name)
links = tuple(reader.iter_links(data, encoding=name))
link = links[0]
self.assertEqual('blah', link)
def test_sitemap_scraper_xml(self):
scraper = SitemapScraper(HTMLParser())
request = Request('http://example.com/sitemap.xml')
response = Response(200, 'OK')
response.body = Body()
with wpull.util.reset_file_offset(response.body):
response.body.write(b'''<?xml version="1.0" encoding="UTF-8"?>
<urlset xmlns="http://www.sitemaps.org/schemas/sitemap/0.9">
<url>
<loc>http://www.example.com/</loc>
<lastmod>2005-01-01</lastmod>
<changefreq>monthly</changefreq>
<priority>0.8</priority>
</url>
</urlset>
''')
scrape_result = scraper.scrape(request, response)
inline_urls = scrape_result.inline_links
linked_urls = scrape_result.linked_links
self.assertEqual({
'http://www.example.com/',
}, linked_urls)
self.assertFalse(inline_urls)
def test_sitemap_scraper_xml_index(self):
scraper = SitemapScraper(HTMLParser())
request = Request('http://example.com/sitemap.xml')
response = Response(200, 'OK')
response.body = Body()
with wpull.util.reset_file_offset(response.body):
response.body.write(b'''<?xml version="1.0" encoding="UTF-8"?>
<sitemapindex
xmlns="http://www.sitemaps.org/schemas/sitemap/0.9">
<sitemap>
<loc>http://www.example.com/sitemap1.xml.gz</loc>
<lastmod>2004-10-01T18:23:17+00:00</lastmod>
</sitemap>
</sitemapindex>
''')
scrape_result = scraper.scrape(request, response)
inline_urls = scrape_result.inline_links
linked_urls = scrape_result.linked_links
self.assertEqual({
'http://www.example.com/sitemap1.xml.gz',
}, linked_urls)
self.assertFalse(inline_urls)
def test_html_scraper_links_base_href(self):
element_walker = ElementWalker(css_scraper=CSSScraper(),
javascript_scraper=JavaScriptScraper())
scraper = HTMLScraper(HTMLParser(), element_walker)
request = Request('http://example.com/')
response = Response(200, 'OK')
response.body = Body()
with wpull.util.reset_file_offset(response.body):
html_file_path = os.path.join(ROOT_PATH, 'testing', 'samples',
'basehref.html')
with open(html_file_path, 'rb') as in_file:
shutil.copyfileobj(in_file, response.body)
scrape_result = scraper.scrape(request, response)
inline_urls = scrape_result.inline_links
linked_urls = scrape_result.linked_links
self.assertEqual('utf-8', scrape_result.encoding)
self.assertEqual(
{
'http://cdn.example.com/stylesheet1.css',
'http://www.example.com/stylesheet2.css',
'http://example.com/a/stylesheet3.css',
'http://example.com/a/dir/image1.png',
'http://example.com/dir/image2.png',
'http://example.net/image3.png',
'http://example.com/dir/image4.png',
}, inline_urls)
self.assertEqual({'http://example.com/a/'}, linked_urls)
def test_html_encoding(self):
html_parser = HTMLParser()
reader = HTMLReader(html_parser)
bom_map = {
'utf_16_le': codecs.BOM_UTF16_LE,
'utf_16_be': codecs.BOM_UTF16_BE,
'utf_32_le': codecs.BOM_UTF32_LE,
'utf_32_be': codecs.BOM_UTF32_BE,
}
for name in CODEC_NAMES:
if name in EBCDIC or name == 'utf_8_sig':
# XXX: we're assuming that all codecs are ASCII backward
# compatable
continue
if name.startswith('utf_16') or name.startswith('utf_32'):
# FIXME: libxml/lxml doesn't like it when we pass in a codec
# name but don't specify the endian but BOM is included
continue
print('->', name)
data = io.BytesIO(bom_map.get(name, b'') + '<img>'.encode(name))
elements = tuple(reader.iter_elements(data, encoding=name))
html_element = elements[0]
self.assertEqual('html', html_element.tag)
def test_html_script_comment(self):
test_string = b'''<script><!-- blah --></script>'''
reader = HTMLReader(HTMLParser())
elements = reader.iter_elements(io.BytesIO(test_string),
encoding='ascii')
elements = tuple(elements)
self.assertTrue(
all(isinstance(element, Element) for element in elements))
def test_sitemap_scraper_invalid_robots(self):
scraper = SitemapScraper(HTMLParser())
request = Request('http://example.com/robots.txt')
response = Response(200, 'OK')
response.body = Body()
with wpull.util.reset_file_offset(response.body):
response.body.write(
b'dsfju3wrji kjasSItemapsdmjfkl wekie;er :Ads fkj3m /Dk')
scrape_result = scraper.scrape(request, response)
inline_urls = scrape_result.inline_links
linked_urls = scrape_result.linked_links
self.assertFalse(linked_urls)
self.assertFalse(inline_urls)
def test_bad_xml(self):
element_walker = ElementWalker(css_scraper=CSSScraper(),
javascript_scraper=JavaScriptScraper())
scraper = HTMLScraper(HTMLParser(), element_walker)
request = Request('http://example.com/')
response = Response(200, 'OK')
response.body = Body()
with wpull.util.reset_file_offset(response.body):
html_file_path = os.path.join(ROOT_PATH, 'testing', 'samples',
'foxstripcomics_bad_xml.html')
with open(html_file_path, 'rb') as in_file:
shutil.copyfileobj(in_file, response.body)
# No crash
scraper.scrape(request, response, link_type=LinkType.html)
def test_sitemap_scraper_robots(self):
scraper = SitemapScraper(HTMLParser())
request = Request('http://example.com/robots.txt')
response = Response(200, 'OK')
response.body = Body()
with wpull.util.reset_file_offset(response.body):
response.body.write(b'Sitemap: http://example.com/sitemap00.xml')
scrape_result = scraper.scrape(request, response)
inline_urls = scrape_result.inline_links
linked_urls = scrape_result.linked_links
self.assertEqual({
'http://example.com/sitemap00.xml',
}, linked_urls)
self.assertFalse(inline_urls)
def test_html_encoding_lxml_name_mismatch(self):
'''It should accept encoding names with underscore.'''
element_walker = ElementWalker(css_scraper=CSSScraper(),
javascript_scraper=JavaScriptScraper())
scraper = HTMLScraper(HTMLParser(), element_walker)
request = Request('http://example.com/')
response = Response(200, '')
response.body = Body()
response.fields['content-type'] = 'text/html; charset=EUC_KR'
with wpull.util.reset_file_offset(response.body):
response.body.write('힖'.encode('euc_kr'))
scrape_info = scraper.scrape(request, response)
self.assertTrue(scrape_info)
self.assertEqual('euc_kr', scrape_info['encoding'])
def test_html_scraper_reject_type(self):
element_walker = ElementWalker(css_scraper=CSSScraper(),
javascript_scraper=JavaScriptScraper())
scraper = HTMLScraper(HTMLParser(), element_walker)
request = Request('http://example.com/')
response = Response(200, 'OK')
response.body = Body()
with wpull.util.reset_file_offset(response.body):
html_file_path = os.path.join(ROOT_PATH, 'testing', 'samples',
'many_urls.html')
with open(html_file_path, 'rb') as in_file:
shutil.copyfileobj(in_file, response.body)
scrape_result = scraper.scrape(request,
response,
link_type=LinkType.css)
self.assertFalse(scrape_result)
def test_html_garbage(self):
element_walker = ElementWalker(css_scraper=CSSScraper(),
javascript_scraper=JavaScriptScraper())
scraper = HTMLScraper(HTMLParser(), element_walker)
request = Request('http://example.com/')
response = Response(200, '')
response.body = Body()
response.fields['content-type'] = 'text/html'
with wpull.util.reset_file_offset(response.body):
response.body.write(
b'\x01\x00\x01\x00l~Z\xff\x0f`y\x80\x00p<\x7f'
b'\xffndo\xff\xff-\x83{d\xec</\xfe\x80\x00\xb4Bo'
b'\x7f\xff\xff\xffV\xc1\xff\x7f\xff7')
scrape_info = scraper.scrape(request, response)
self.assertTrue(scrape_info)
def test_html_parse_doctype(self):
html_parser = HTMLParser()
self.assertIn(
'html',
html_parser.parse_doctype(
io.BytesIO(b'<!DOCTYPE HTML><html></html>')))
self.assertIn(
'XHTML',
html_parser.parse_doctype(
io.BytesIO(b'''
<!DOCTYPE html PUBLIC
"-//W3C//DTD XHTML 1.0 Transitional//EN"
"http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
<html></html>
''')))
self.assertFalse(html_parser.parse_doctype(
io.BytesIO(b'hello world!')))
self.assertFalse(html_parser.parse_doctype(io.BytesIO(b'')))
self.assertFalse(html_parser.parse_doctype(io.BytesIO(b'\x00')))
self.assertFalse(html_parser.parse_doctype(io.BytesIO(b'A\xfe')))
def test_html_serious_bad_encoding(self):
element_walker = ElementWalker(css_scraper=CSSScraper(),
javascript_scraper=JavaScriptScraper())
scraper = HTMLScraper(HTMLParser(),
element_walker,
encoding_override='utf8')
request = Request('http://example.com/')
response = Response(200, '')
response.body = Body()
response.fields['content-type'] = 'text/html; charset=utf8'
with wpull.util.reset_file_offset(response.body):
html_file_path = os.path.join(ROOT_PATH, 'testing', 'samples',
'xkcd_1_evil.html')
with open(html_file_path, 'rb') as in_file:
shutil.copyfileobj(in_file, response.body)
scrape_info = scraper.scrape(request, response)
self.assertTrue(scrape_info)
def test_html_not_quite_charset(self):
element_walker = ElementWalker(css_scraper=CSSScraper(),
javascript_scraper=JavaScriptScraper())
scraper = HTMLScraper(HTMLParser(), element_walker)
request = Request('http://example.com/')
response = Response(200, '')
response.body = Body()
with wpull.util.reset_file_offset(response.body):
html_file_path = os.path.join(ROOT_PATH, 'testing', 'samples',
'videogame_top.htm')
with open(html_file_path, 'rb') as in_file:
shutil.copyfileobj(in_file, response.body)
scrape_result = scraper.scrape(request, response)
inline_urls = scrape_result.inline_links
linked_urls = scrape_result.linked_links
self.assertIn('http://example.com/copyright_2001_2006_rtype.gif',
inline_urls)
self.assertIn('http://www.geocities.jp/gamehouse_grindcrusher/',
linked_urls)
def test_html_wrong_charset(self):
element_walker = ElementWalker(css_scraper=CSSScraper(),
javascript_scraper=JavaScriptScraper())
scraper = HTMLScraper(HTMLParser(), element_walker)
request = Request('http://example.com/')
response = Response(200, '')
response.body = Body()
with wpull.util.reset_file_offset(response.body):
html_file_path = os.path.join(ROOT_PATH, 'testing', 'samples',
'kcna.html')
with open(html_file_path, 'rb') as in_file:
shutil.copyfileobj(in_file, response.body)
scrape_result = scraper.scrape(request, response)
inline_urls = scrape_result.inline_links
linked_urls = scrape_result.linked_links
self.assertEqual('utf-16-le', scrape_result.encoding)
self.assertEqual(
{
'http://example.com/utm/__utm.js',
'http://example.com/Knewskage.gif',
'http://example.com/Lline.gif',
'http://example.com/Sline.gif',
'http://example.com/korean01.gif',
'http://example.com/korean02.gif',
'http://example.com/english01.gif',
'http://example.com/english02.gif',
'http://example.com/Tongsinkage.gif',
'http://example.com/Knewskage.gif',
}, inline_urls)
self.assertEqual(
{
'http://example.com/index-k.htm',
'http://example.com/index-e.htm',
}, linked_urls)
def test_html_layout(self):
html_parser = HTMLParser()
reader = HTMLReader(html_parser)
elements = tuple(
reader.iter_elements(io.BytesIO(b'''
<html>
<head>
<title>hi</title>
</head>
<body>
<img>
</body>
</html>'''),
encoding='ascii'))
print(elements)
self.assertEqual('html', elements[0].tag)
self.assertEqual('head', elements[1].tag)
self.assertEqual('title', elements[2].tag)
self.assertEqual('body', elements[3].tag)
self.assertEqual('img', elements[4].tag)
def test_html_krokozyabry(self):
element_walker = ElementWalker(css_scraper=CSSScraper(),
javascript_scraper=JavaScriptScraper())
scraper = HTMLScraper(HTMLParser(), element_walker)
request = Request('http://example.com/')
response = Response(200, '')
response.body = Body()
response.fields['content-type'] = 'text/html; charset=KOI8-R'
with wpull.util.reset_file_offset(response.body):
html_file_path = os.path.join(ROOT_PATH, 'testing', 'samples',
'krokozyabry.html')
with open(html_file_path, 'rb') as in_file:
shutil.copyfileobj(in_file, response.body)
scrape_result = scraper.scrape(request, response)
inline_urls = scrape_result.inline_links
linked_urls = scrape_result.linked_links
self.assertEqual('koi8-r', scrape_result.encoding)
self.assertEqual(set(), inline_urls)
self.assertEqual({'http://example.com/Кракозябры'}, linked_urls)
def test_xhtml_invalid(self):
element_walker = ElementWalker(css_scraper=CSSScraper(),
javascript_scraper=JavaScriptScraper())
scraper = HTMLScraper(HTMLParser(), element_walker)
request = Request('http://example.com/')
response = Response(200, '')
response.body = Body()
with wpull.util.reset_file_offset(response.body):
html_file_path = os.path.join(ROOT_PATH, 'testing', 'samples',
'xhtml_invalid.html')
with open(html_file_path, 'rb') as in_file:
shutil.copyfileobj(in_file, response.body)
scrape_result = scraper.scrape(request, response)
inline_urls = scrape_result.inline_links
linked_urls = scrape_result.linked_links
self.assertEqual(
{
'http://example.com/image.png',
'http://example.com/script.js',
}, inline_urls)
self.assertEqual({'http://example.com/link'}, linked_urls)
def test_html_early_html(self):
reader = HTMLReader(HTMLParser())
for test_string in [
b'''<!DOCTYPE HTML><html></html><img>''',
b'''<html></html><img>''',
b'''<!DOCTYPE HTML><img><html></html>''',
b'''<img><html></html>''',
b'''<!DOCTYPE HTML>
<html><body></body></html><p><img>''',
b'''
<html><body></body></html><p><img>''',
b'''
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN"
"http://www.w3.org/TR/html4/loose.dtd">
<html>
<head>
<title>Download</title>
</head>
<body>
<br />
</body>
</html>
<pre><img></pre>
''',
b'''
<html>
<head>
<title>Download</title>
</head>
<body>
<br />
</body>
</html>
<pre><img></pre>
''',
b'''
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN"
"http://www.w3.org/TR/html4/loose.dtd">
<html>
<body>
<br />
</body>
<head>
<title>Download</title>
</head>
</html>
<pre><img></pre>
''',
b'''
<html>
<body>
<br />
</body>
<head>
<title>Download</title>
</head>
</html>
<pre><img></pre>
''',
]:
elements = []
print()
print('a' * 10)
print(test_string)
for element in reader.iter_elements(io.BytesIO(test_string),
encoding='ascii'):
if isinstance(element, Element):
print(element)
elements.append(element)
element_tags = tuple(element.tag for element in elements)
self.assertIn('img', element_tags)
def test_html_scraper_links(self):
element_walker = ElementWalker(css_scraper=CSSScraper(),
javascript_scraper=JavaScriptScraper())
scraper = HTMLScraper(HTMLParser(), element_walker)
request = Request('http://example.com/')
response = Response(200, 'OK')
response.body = Body()
response.fields['Refresh'] = '3; url=header_refresh.html'
with wpull.util.reset_file_offset(response.body):
html_file_path = os.path.join(ROOT_PATH, 'testing', 'samples',
'many_urls.html')
with open(html_file_path, 'rb') as in_file:
shutil.copyfileobj(in_file, response.body)
scrape_result = scraper.scrape(request, response)
inline_urls = scrape_result.inline_links
linked_urls = scrape_result.linked_links
self.assertEqual('utf-8', scrape_result.encoding)
self.assertEqual(
frozenset({
'http://example.com/style_import_url.css',
'http://example.com/style_import_quote_url.css',
'http://example.com/style_single_quote_import.css',
'http://example.com/style_double_quote_import.css',
'http://example.com/bg.png',
'http://example.com/link_href.css',
'http://example.com/script.js',
'http://example.com/body_background.png',
'http://example.com/images/table_background.png',
'http://example.com/images/td_background.png',
'http://example.com/images/th_background.png',
'http://example.com/style_url1.png',
'http://example.com/style_url2.png',
'http://example.com/applet/', # returned by lxml
'http://example.com/applet/applet_code.class',
'http://example.com/applet/applet_src.class',
'http://example.com/bgsound.mid',
'http://example.com/audio_src.wav',
'http://example.com/audio_poster.jpeg',
'http://example.net/source_src.wav',
'http://example.com/video_src.webm',
'http://example.com/video_poster.jpeg',
'http://example.net/track_src.vtt',
'http://example.net/source_src.webm',
'http://example.com/embed_src.mov',
'http://example.com/fig_src.png',
'http://example.com/frame_src.html',
'http://example.com/iframe_src.html',
'http://example.com/img_href.png',
'http://example.com/img_lowsrc.png',
'http://example.com/img_src.png',
'http://example.com/img_data.png',
'http://example.com/img_srcset_1.jpeg',
'http://example.com/img_srcset_2.jpeg',
'http://example.com/img_srcset_3.jpeg',
'http://example.com/input_src.png',
'http://example.com/layer_src.png',
'http://example.com/object/', # returned by lxml
'http://example.com/object/object_data.swf',
'http://example.com/object/object_archive.dat',
'mailto:internet',
'object_not_url_codebase',
'http://example.com/param_ref_value.php',
'http://example.com/overlay_src.html',
'http://example.com/script_variable.png',
}),
inline_urls)
self.assertEqual(
frozenset({
'http://example.com/og_image.png',
'http://example.com/og_url.html',
'http://example.com/og_audio.mp3',
'http://example.com/og_video.webm',
'http://example.com/twitter_image.png',
'http://example.com/twitter_image0.png',
'http://example.com/twitter_image1.png',
'http://example.com/twitter_image2.png',
'http://example.com/twitter_image3.png',
'http://example.com/twitter_player.html',
'http://example.com/twitter_stream.mp4',
'http://example.net/soup.html',
'http://example.com/a_href.html',
'http://example.com/area_href.html',
'http://example.com/frame_src.html',
'http://example.com/embed_href.html',
'http://example.com/embed_src.mov',
'http://example.com/form_action.html',
'http://example.com/iframe_src.html',
'http://example.com/layer_src.png',
'http://example.com/overlay_src.html',
'ftp://ftp.protocol.invalid/',
'mailto:[email protected]',
'http://a-double-slash.example',
'http://example.com/header_refresh.html',
'https://[2001:db8:85a3:8d3:1319:8a2e:370:7348]:8080/ipv6',
'http://example.com/document_write.html',
'http://example.com/http_document_write.html',
'http://example.com/http_document_write2.html',
'http://example.com/http document write.html',
'http://example.com/script_variable.html',
'http://example.com/http_script_variable.html',
'https://example.com/https_script_variable.html',
'ftp://example.com/ftp_script_variable.html',
'http://example.com/end_dir_script_variable/',
'http://example.com/start_dir_script_variable',
'http://example.com/../relative_dir_script_variable'
if sys.version_info <
(3, 5) else 'http://example.com/relative_dir_script_variable',
'http://example.com/script_json.html',
'http://example.com/http_script_json.html?a=b',
'http://example.com/a_javascript_link.html',
'http://example.com/a_onclick_link.html',
}), linked_urls)
for url in inline_urls | linked_urls:
self.assertIsInstance(url, str)
